Sunteți pe pagina 1din 19

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.

1, January 2015

LEARNING STRATEGY WITH GROUPS ON PAGE

BASED STUDENTS' PROFILES
San San Tint1 and Ei Ei Nyunt2
1

Department of Research and Development II, University of Computer Studies,

Mandalay, Myanmar
2
Master of Computer Science, University of computer Studies Mandalay, Myanmar

Abstract
Most of students desire to know about their knowledge level to perfect their exams. In learning environment
the fields of study overwhelm on page with collaboration or cooperation. Students can do their exercises
either individually or collaboratively with their peers. The system provides the guidelines for students'
learning system about interest fields as Java in this system. Especially the system feedbacks information
with each others because of sharing e mail and list of them.

Keywords

1. INTRODUCTION
Thinking of participants is the first step of collaborative learning system. In this step the system
used the student attributes or properties from their profiles. The properties and the values of the
properties are specified by the system. The think step and pair step of the Think-Pair-Share (TPS)
Strategy are measured by the clustering method such as K-means method [1-5].
Data clustering is a data exploration technique that allows objects with similar characteristics to
be grouped together in order to facilitate their further processing. In the system clustering method
is used to produce the groups of student participants according to their properties as shown in
Table 1. The objects are the students and the initial number of cluster is specified with the random
number of students [4], [6].
Cluster analysis is a formal study of methods for understanding and algorithm for learning. Kmean is the first choice for clustering with an initial number of clusters. K-mean algorithm is
most widely used algorithm in data mining applications. It is a simple, scalable, easily
understandable and can be adopted to deal with high dimensional data. A distance measuring
function is used to measure the similarity among objects, in such a way that more similar objects
have lower dissimilarity value. Several distance measures can be employed for clustering tasks.
The K-Mean algorithm finds partitions with distance measuring function in the cluster which is
minimized [7-9].

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015

2. BACKGROUND THEORY
2.1. Think-Pair-Share (TPS) Strategy
Think-Pair-Share is a relatively low-risk and short collaborative learning technique, and is suited
for instructors and students who are new to collaborative learning.
Think-Pair-Share technique in education is also about:





The purposes of this technique are to process information, having a communication and develop
thinking among students. This strategy helps students become active participants in learning and
can include writing as a way of organizing thoughts generated from discussions [6].

2.2. Think-Pair-Share Technique Role

The teacher decides upon the text to be read and develops the set of questions or prompts that
target key content concepts. The teacher then describes the purpose of the strategy and provides
guidelines for discussions. As with all strategy instruction, teachers should model the procedure
to ensure that students understand how to use the strategy. Teachers should monitor and support
students as they work. In the system the Think-Pair-Share technique will be applied as following:
Thinking: each student thinking of his/her profile attributes and answering the given questions.
Pair: Pair the students grade and the K-means clustering result which is used students profile.
The students of grades (A-C) will be specified by using his/her exam result.
Share: Share the students list that contains not only Grade A and Grade B with their results but
also their email and original group number to communicate each other.
When the system basic level step is finished, both thinking and first part of pairing of the students
who get other Grade can retry the questions next time [3].

2.3. The Main properties of Objects (7 attributes)

The objects of the system are students data which are specified by the system. The main
attributes of student information are specified as objects properties to calculate the similarity
among objects. The attributes are specified by the collaborative learning system and are
calculated as k-means objects to determine the group of the participants. The attributes and values
of the clustering categories are specified as following Table 1. These all attributes are belonging
to the student objects which are specified by the students data of their profile. The system is
defined the students to fill their data which is belong to the following properties and values. The
students information and exam results are calculated to apply the TPS theories in this system.

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015
Table .1 Attributes and Values for Student Objects

Attribute Name
Education
Occupation
Math Skill
Physic Skill
Programming
Skill
10th
standard
Passed Year
Interest
in
subject

Value 1
Information Technology
Student
Normal
Normal

Value 2
E.P Engineering

Value 3
Computer Technology
Distinction
Distinction

Learner

Developer

Expert

2006-2013(after)

1998-2005

1990-1997

OOP

Networking

Web Development

These attributes are specified for the calculation of clustering of the student objects. There are
three or five clusters to group the students. The students attributes are calculated by the
Euclidean distance function to determine the groups of students [11].

2.4. Distance Measuring Function for K-Means Clustering

The system specifies the random objects or students by using Rnd function on attributes. And get
top three or five objects to specify the centroids of initial clusters. Then the system calculates the
members of clusters by using the Euclidean distance function which is shown as following:
(1)
The mean for a cluster is:

1
C j

xiC

xi
j

(2)

Where, X 1 = ( x11 , x12 ,..., x1n )

X 2 = ( x21 , x22 ,..., x2 n )
C j = number of data points in cluster C j
2.4.1. K-Means Clustering Algorithm for the System
Input:
Let k=5 is the number of clusters to partition.
D is a database containing n objects:(n is the number of students attributes from profile).
Output:
A set of k clusters: ( including class of attributes with their members objects.)
Method:
arbitrarily choose k objects from D as the initial cluster centers;
Let Ct =new centroids ( t=1,2,3,,k)
3

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015

xi=arbitrarily students attributes(i=1,2,3,,k)

xj=all students attributes
(j=1,2,3,,n)
Dt =data set of distance values (t=1,2,3,,k)
Initialize xi=1 , xj=1;
3)
repeat j
repeat i
d(xi ,xj) = || xi xj ||2
increase i
until i <= k
choose minimum distance value and assign to Dt.
increase t
increase j
until j <= n
Compare Dt with minimize pair value and reassign each object to the cluster to which the object
is the most similar based on the means value of the objects in the cluster (Ct, where t=1, 2,
3,,k).Update the cluster means; i.e., calculate the mean value of the objects for each cluster
until no change.
2.4.2. Roles of Student Profiles
In the clustering method of the system, the role of students profiles is to calculate the groups of
students. The groups of students are specified by the clustering method. Then the students answer
the exam questions which are become the input of next step of the collaborative learning of the
system. The next step is the calculation of paring students according to the result of their exam
marks and their grades. The inputs of the first step are as shown in the following Table 2.
Table 2. Example profile table of the Students

Stude
nt
Obj
a1
a2
a3
a4
b1
b2
b3
c1
c2
c3

Education

Occupation

Math
Skill

Physic
Skill

Program
Skill

1
3
2
1
1
2
3
2
1
3

2
1
1
1
3
1
2
3
1
3

1
2
2
1
2
2
2
1
3
3

2
1
2
1
1
1
1
3
1
2

2
1
1
1
2
1
3
1
2
3

Education
Occupation
Math Skill
Physic Skill
Program Skill
10th Standard Passed Year
Interested in Subject

1 = IT
1 = Student
1 = Normal
1 = Normal
1 = Learner
1 = 2006-2013
later
1 = OOP

2 = EP
2 = Developer
2 = 1998-2005
2 = Networking

10th Std
Passed
Year
2
1
1
1
2
1
2
3
1
3

Interested
in Subject
3
1
1
2
3
1
3
2
2
2

3 = Computer
3 = Distinction
3 = Distinction
3 = Expert
3= 1990-1997
3=Web
4

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015

In K-means method, the number of clusters must be specified as the initial clusters centroid.
Therefore the example objects are specified from the student profile table as the initial clusters
centroid are as shown in the following Table 3. The initial cluster centroids are calculated with
other objects in the student profile table by substitution of the Euclidean distance function.
Table 3. Objects from the Profile as initial Cluster

Obj

Education

Occupation

Math
Skill

Physic
Skill

Program
Skill

a1
b2
c3

1
2
3

2
1
3

1
2
3

2
1
2

2
1
3

10th Std
Passed
Year
2
1
3

Interested
in Subject
3
1
2

Table 4. Centroids for Second Iteration

Interested
in Subject

1.7
1.2

10th Std
Passed
Year
2.3
1

2.5

2.5

Student
Obj

Education

Occupation

Math
Skill

Physic
Skill

Program
Skill

a1 , b1, c1
a2 ,a3 ,a4
,b2 ,c2
b3 , c3

1.3
1.8

2.7
1

1.3
2

2
1.2

2.5

2.5

1.5

2.7
1.4

Step 6
The centroid from second iteration is no change. The k-means algorithm is terminated. The
cluster is illustrated with set graph as shown in fig.1.

2.5. Distance Measuring for New Student

The distance Measure for new student is calculated with Table 5 cetroids which are derived from
k-mean distance measuring function. The derivation of new students group is as shown in the
following Table 6.

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015
Table 5. New Students Profile

Dist (a1 b1c1 , d )

= 3.967
Dist (a 2 a3 a 4 b2 c 2 , d ) = 1.5
Dist (b3 c3 , d )
= 3.354

2.6. Pair and Sharing of Participants

The pair and sharing of participants is the final step of the system. The system thinks the
participants or students skills both calculating the k-mean clustering method and examination
method. The system decides the pair of participants with the grade of advance level students
marks. Then the system shares the similar upper levels of students skill. The students will answer
two steps of exam with their groups. The finally the students are shared among the upper two
levels of groups with their examination information.

Marks
>=8
7
<4

In the sharing stage of students, the system gives the students information such as email
addresses each other. The students will go on sharing and learning the sense or assuming on the
relative subject with their mail addresses.

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015

3. IMPLEMENTATION OF THE SYSTEM

The two main parts of the implementation system have explained as shown in the following
figures. The figures demonstrate the collaborative system how to cluster the profiles and hold the
examinations.

Figure 3. Main Page for Administrator

7

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015

The Admin Role Page, Figure 5 consists of seven links such as Arrange Exam Date, Arrange
Groups, Insert Questions, View Marks and View Students. Log Out! and Home links are used to

Figure 7. Error Page for Arrange Exam Date

8

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015

Figure 8. The New Arrange Exam Date

Above Figure 6, 7, and 8 illustrate the specification of New Exam Date. According to Figure 6
the admin needs to type Exam ID and Exam ID fields. And then press Add Date button. In Figure
7 the system replies that the exam date is pass over or impossible date to specify for the
examination. In Figure 8 the Exam Date is successfully specified by the administrator and then
the system displays the new Exam Date with old list.

Figure 9. Arrange Group with Initial Group Numbers

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015
Figure 10. Arrange Groups with Initial Cluster Number

Figure 12. Arrange Groups with Students Profiles List

10

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015

Figure 13. Error Page for Arrange Groups

From Figure 6 to 12 are shown the Arrange Groups. The system shows the arrange groups
numbers for initial clusters as shown in Figure 10. Figure 11 and 12 show the three random
students profiles to calculate the groups of students who are signed up the system. If the exam is
hold on today, then the administrator cannot specify the groups. Figure 12 shows the students
profile records with their groups. The error message as in Figure 13 will be shown.
3.6. Insert Question Page

Figure 14. Question Entry Form

11

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015

The admin can add question by using Question Entry Form as shown in Figure 14 and the
administrator fills this form and press Add button. Then the system saves the question to the
database.

3.7. View Marks Page

Figure 15. View the Student List With their exam Marks

Figure 16. Error Page for View Marks

The administrator can view the students list with their exam marks and groups as shown in the
Figure 15. When the students are grouped next time and the exam date has been specified or the
new examination is not hold yet, the error page for View Marks will be displayed as shown in
Figure 16.

12

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015

Figure 17. Student Main Page

Student Main Page as shown in Figure 17 represents the students or participant of the
Collaborative Learning System. The student can enter into the system by using the Student Login

Figure 18 shows the login page for student by using their names and passwords. If there is no
exam for the student today, the system shows Message as shown in Figure 19. Moreover, the
exam date is not specified, then the system will display the old shared groups list with student as
in Figure 20.

13

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015

Figure 20. Student Group List Page

But the system shows the past student grades with their groups list to the login student. If the
student has completed the past exam, and the grade is high then he/she can see the same students
groups. If the student is new for this system, he/she must fill the form as shown in Figure 21. If
today is the exam date, the new student cannot enter the exam pages as in Figure 22.

14

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015

3.10. Grouping and Pairing Students

In this part the student can see the examination entrance page to participate the exam with their
groups. Figure 23 shows the entrance page for student to enter the examination. At first, the
student can answer basic level of exam and can send his/her answers by pressing the submit
button as in Figure 24. Then the system display basic level marks and grade of student as shown
in Figure 25. If the student's grade does not reached at specified level, the student must answers
the basic level again. If the students log out at the end of basic level, the system shows the basic
grade. When the student re-enters again to the system, the system shows the message as shown in
Figure 26. If the student's the grade reaches at specified level, the student can answer the advance
level as the next step. After finishing the exam, the student can see his/her grade with his/her
marks. If the student reached at the specified level, he/she can see the shared list with his/her
name as shown in Figure 28. But the student did not reach at specified level, his/her name does
not appear in upper list.
15

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015

Figure 25. Basic Level of Students Marks, Grade and Group

16

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015

Figure 28. Error Entrance of Student who answered all questions

17

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015

Figure 29. Student Grade with Marks

In Figure 28, there is a Error Entrance of Student who answered all questions. If the student had
answered all questions and passed level, he/she can see the message as in Figure 28 when he/she
enters to the system again. In Figure 29, the student can see only the groups of students who get

4. CONCLUSION
This system aims at the students in order to promote active learning in computer based learning
environment. A well-known collaborative learning technique, the Think-Pair-Share is applied
because it has simplicity and suitability to be implemented in a collaborative learning
environment. This system provides the benefits to specify the grades and group of the students by
using K-mean clustering algorithm and also improves the students' learning without difficulties to
find out their interest fields.

ACKNOWLEDGEMENTS
Our heartfelt thanks go to all people, who support us at the University of Computer Studies,
Mandalay, Myanmar. This paper is dedicated to our parents. Our special thanks go to all
respectable persons who support for valuable suggestion in this paper.

REFERENCES
[1]

[2]
[3]
[4]
[5]

N. A. N. Azlina, "CETLs : Supporting Collaborative Activities Among Students and Teachers

Through the Use of Think-Pair-Share Techniques", IJCSI International Journal of Computer Science
Issues, Vol. 7, Issue 5, September 2010, ISSN (Online): 1694-0814, www. IJCSI. org.
Grand Rapids Community College Center, "Ten Techniques For Energizing Your Classroom
Discussions for Teaching and Learning ", [On-line] http://web.grcc.cc.mi.us.
Schreyer Institute for Teaching Excellence, Penn State, 301 Rider Building II, University Park, PA
16802, www.schreyerinstitute.psu.edu, 2007.
SI Showcase, "The Basic Collaborative Learning Techniques", Supplemental Instruction Iowa State
University, 1060 Hixson-Lied Student Success Center, 294-6624, www.si.iastate.edu.
E. Barkley, K. P. Cross and C. Major, "Collaboration learning techniques", San Francisco, CA:
Jossey-Bass, 2005.
18

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.1, January 2015
[6]

J. M. Tighe and F. T. Lyman, Cueing Thinking in the Classroom: The Promise of TheoryEmbedded Tools, Educational Leadership, 1988, Vol. 45, pp. 18-24.
[7] T. Yerigan, Getting Active In The Classroom., Journal of College Teaching and Learning, Vol. 5,
Issue 6, 2008, pp. 19-24.
[8] C. Opitz, and W. L. Bowman, " ", Elementary School, Anchorage School District, 2008.
[9] P. Berkhin, Knoll and S. Jose, "Survey of Clustering Data Mining Techniques", Pavel Berkhin
Accrue Software, Inc.
[10] P. Vora and B. Oza, A Survey on K-mean Clustering and Particle Swarm Optimization",
International Journal of Science and Modern Engineering (IJISME) ISSN: 2319-6386, Volume-1,
Issue-3, February, 2013.
[11]
N. Kaur, J. K. Sahiwal and N. Kaur, Efficient K-MEANS Clustering Algorithm using Ranking
Method in Data Mining, ISSN: 2278 1323, International Journal of Advanced Research in Computer
Engineering & Technology, Volume 1, Issue 3, May 2012.

Authors
She is Associate Professor, Head of Department of Research and Development II in
University of Computer Studies, Mandalay, Myanmar. Her research areas include
Information Retrieval, Cryptography and Network Security, Web Mining and Networking.
She received her B.Sc. (Physics), M.Sc.(Physics) from Yangon University, Myanmar and
M.A.Sc.(Computer Engineering) and Ph.D.( Information Technology) from University of
Computer Studies, Yangon, Myanmar.Author studied computer science at the University of Computer
Studies, lasho, Myanmar where she received her B.C.Sc Degree in 2011. She received B.C.Sc(Hons:) in
computer science from the University of Computer Studies, Lasho, Myanmar in 2012. Since 2012, Author
has studied computer science at the University of Computer Studies, Mandalay, Myanmar where her
primary interests include web mining, graph clustering, grouping and web log analysis.

19